NEWS
LIFESTYLE
FUNNY
WHOLESOME
INSPIRING
ANIMALS
RELATIONSHIPS
PARENTING
WORK
SCIENCE AND NATURE
About Us Contact Us Privacy Policy
SCOOP UPWORTHY is part of
GOOD Worldwide Inc. publishing
family.
© GOOD Worldwide Inc. All Rights Reserved.

Teen invents low-cost machine learning tool to detect elephant poachers in real time

According to the teen, the software is four times more accurate than existing state-of-the-art detection methods.

Teen invents low-cost machine learning tool to detect elephant poachers in real time
Cover Image Source: Society for Science

Anika Puri's fight against poaching began four years ago during a visit to India with her family. The teen was shocked when she came across a market in Mumbai filled with rows of ivory jewelry and statues. "I was quite taken aback," the 17-year-old from Chappaqua, New York, told Smithsonian Magazine. "Because I always thought, 'well, poaching is illegal, how come it really is still such a big issue?'" Intrigued, Puri did some research on the matter and was stunned when she discovered Africa's forest elephant population had declined by about 62% between the years 2002 and 2011. The numbers have continued to drop in the years that followed.

Image Source: Society for Science
Image Source: Society for Science

As a wildlife lover, Puri was determined to do something to help. She learned that although drones are currently used to detect and capture images of poachers, they're not very accurate. The teenager pored over videos of elephants and humans and noticed how vastly the two differed in the way they move. "I realized that we could use this disparity between these two movement patterns in order to actually increase the detection accuracy of potential poachers," she explained. Armed with this observation, Puri set about creating ElSa (an acronym for elephant savior), a low-cost prototype of a machine-learning-driven software that analyzes movement patterns in thermal infrared videos of humans and elephants.

Image Source: Society for Science
Image Source: Society for Science

Puri reached out to Elizabeth Bondi-Kelly, a Harvard computer scientist who was working on a wildlife conservation project using drones and machine learning, with her idea and the computer scientist agreed to become her mentor for the project. This proved to be a valuable connection as it gave the teen access to the Benchmarking IR Dataset for Surveillance with Aerial Intelligence (BIRDSAI), a dataset collected by Bondi-Kelly and her colleagues using a thermal infrared camera attached to an unmanned aerial vehicle (UAV) in multiple protected areas in Africa. From the data, Puri picked out 516 time series extracted from videos that captured humans or elephants in motion.

She then used a machine learning algorithm on these movement patterns to train a model to classify a figure as either an elephant or a human on the basis of its speed, group size, turning radius, number of turns and other patterns. Puri used 372 series—300 elephant movements and 72 human movements—for this purpose and the remaining 144 in the testing phase so as to give her model data it hadn't seen before. Her model was able to detect humans with more than 90% accuracy.

Image Source: Society for Science
Image Source: Society for Science

According to Puri, the software—which she developed over the course of two years—is four times more accurate than existing state-of-the-art detection methods. It also eliminates the need for costly high-resolution thermal cameras. The teen submitted her project to this year's Regeneron International Science and Engineering Fair—the world's largest international pre-college STEM competition—and won the Peggy Scripps Award for Science Communication, as well as, a top award in the competition's earth and environmental sciences category.

"It's really amazing just to see all these kids coming together. And for the same purpose—enjoying science and doing research," Puri said of the experience. "I was honored just to be on that stage." She revealed that she first learned about the capabilities of artificial intelligence soon after ninth grade when she was selected to attend Stanford A.I. Lab’s summer program. "Initially, my enthusiasm for artificial intelligence was based off of this limitless possibility for social good," Puri explained. However, she soon realized that since data is collected and analyzed by humans, it contains human biases, and by extension, so does A.I.

Image Source: Society for Science
Image Source: Society for Science

"It really has the capability to reinforce some of the worst aspects of our society," she said. "What I really realized from this is how important it is that women, people of color, all sorts of minorities in the field of technology are at the forefront of this kind of groundbreaking technology." According to Jasper Eikelboom, an ecologist at Wageningen University in the Netherlands who is designing a system to detect poachers using GPS trackers on animals, Puri's software is "quite commendable." 

"It's quite remarkable that a high school student has been able to do something like this," he said. "Not only the research and the analysis but also... being able to implement it in the prototypes." However, Eikelboom cautions that ElSa still needs to be tested on raw video footage to see how well it can detect poachers in real life since the model was tested using figures already determined as either human or elephant.

Image Source: Society for Science
Image Source: Society for Science

Puri is set to attend the Massachusetts Institute of Technology this fall, where she will study electrical engineering and computer science. She also has plans to expand her movement pattern research into other endangered animals. The youngster hopes to begin implementing her software in national parks in Africa—including South Africa's Kruger National Park—after she starts college. "Research isn't a straight line," Puri said. "That has made me more resourceful. It also helped me develop into a more innovative thinker. You learn along the way."



 

More Stories on Scoop